Development of a fiber-guided laser ultrasonic system resilient to high temperature and gamma radiation for nuclear power plant pipe monitoring This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2013 Meas. Sci. Technol. 24 085003 (http://iopscience.iop.org/0957-0233/24/8/085003) Download details: IP Address: 143.248.122.134 The article was downloaded on 04/07/2013 at 01:15 Please note that terms and conditions apply. View the table of contents for this issue, or go to the journal homepage for more Home Search Collections Journals About Contact us My IOPscience
Transcript
Development of a fiber-guided laser ultrasonic system resilient to high temperature and
gamma radiation for nuclear power plant pipe monitoring
This article has been downloaded from IOPscience. Please scroll down to see the full text article.
2013 Meas. Sci. Technol. 24 085003
(http://iopscience.iop.org/0957-0233/24/8/085003)
Download details:
IP Address: 143.248.122.134
The article was downloaded on 04/07/2013 at 01:15
Please note that terms and conditions apply.
View the table of contents for this issue, or go to the journal homepage for more
Home Search Collections Journals About Contact us My IOPscience
Development of a fiber-guided laserultrasonic system resilient to hightemperature and gamma radiation fornuclear power plant pipe monitoringJinyeol Yang, Hyeonseok Lee, Hyung Jin Lim, Nakhyeon Kim,Hwasoo Yeo and Hoon Sohn1
Department of Civil and Environmental Engineering, Korea Advanced Institute of Science andTechnology, Daejeon 305-701, Korea
Received 26 October 2012, in final form 15 May 2013Published 2 July 2013Online at stacks.iop.org/MST/24/085003
AbstractThis study develops an embeddable optical fiber-guided laser ultrasonic system for structuralhealth monitoring (SHM) of pipelines exposed to high temperature and gamma radiationinside nuclear power plants (NPPs). Recently, noncontact laser ultrasonics is gainingpopularity among the SHM community because of its advantageous characteristics such as (a)scanning capability, (b) immunity against electromagnetic interference (EMI) and (c)applicability to high-temperature surfaces. However, its application to NPP pipelines has beenhampered because pipes inside NPPs are often covered by insulators and/or target surfaces arenot easily accessible. To overcome this problem, this study designs embeddable optical fibersand fixtures so that laser beams used for ultrasonic inspection can be transmitted between thelaser sources and the target pipe. For guided-wave generation, an Nd:Yag pulsed laser coupledwith an optical fiber is used. A high-power pulsed laser beam is guided through the opticalfiber onto a target structure. Based on the principle of laser interferometry, the correspondingresponse is measured using a different type of laser beam guided by another optical fiber. Alldevices are especially designed to sustain high temperature and gamma radiation. Therobustness/resilience of the proposed measurement system installed on a stainless steel pipespecimen has been experimentally verified by exposing the specimen to high temperature ofup to 350 ◦C and optical fibers to gamma radiation of up to 125 kGy (20 kGy h−1).
Keywords: pipelines, structural health monitoring (SHM), nuclear power plants (NPPs), laserultrasonics, high temperature, gamma radiation
(Some figures may appear in colour only in the online journal)
1. Introduction
To meet fast-growing future power demands, nuclear energyhas drawn much attention as one of the promising alternativeenergy sources to conventional ones (US Energy InformationAdministration 2008). At the same time, there are growingconcerns about the safety of aging nuclear power plants
1 Author to whom any correspondence should be addressed.
(NPPs). Since the dawn of commercial nuclear power in the1950s, there have been quite a few accidents at NPP sitesaround the world. The most notable accidents include thoseat Three Mile Island (United States, 1979; economic loss:USD 2400M), Chernobyl (Soviet Union, 1986; economic andhuman loss: USD 6700M and 4056 deaths), and, more recently,Fukushima (Japan, 2011; economic loss: USD 2400M). Inresponse to the growing number of NPP disasters, nuclearregulatory authorities in many countries have tightened their
Figure 1. A schematic overview of the proposed optical fiber-guided laser ultrasonic system.
safety measures and established strict inspection criteria foroperational NPPs (American Nuclear Society 2011).
The common and well-established practice for NPPinspection is nondestructive testing (NDT). NDT is acollection of techniques that allow effective localizationand quantification of incipient defects (Carts 1995, Blitzand Simpson 1996). However, there are several drawbacksassociated with NDT. First, some of the current NDTtechniques require periodic overhaul of an NPP facilityreducing power production efficiency. Second, the NDTinspection can be performed only by certified engineers,making it labor-intensive, expensive and time consuming.Finally, there are several critical points that cannot be easilyaccessed by the current NDT techniques. For example, manypipes in NPPs are either covered with insulation materials orburied underground (Adams 2007).
In order to complement existing NDT techniques, theindustry would like to adopt structural health monitoring(SHM) technology to NPP monitoring and management.Compared to NDT, SHM techniques can provide the followingpotential benefits: (1) automated and continuous monitoringof NPP facilities, (2) reduction in downtime, costs and timeassociated with NPP inspection and (3) monitoring of criticalspots, which have been difficult to inspect using conventionalNDT techniques (Sohn et al 2003, Inman 2005). In spiteof these potential benefits, the applications of SHM to NPPfacilities have been partially hampered by the fact thatconventional sensors needed for online SHM cannot surviveharsh operational conditions of NPPs (Hashemian 2010).
This study develops a fiber-guided laser ultrasonicmeasurement system specifically designed for NPP pipelinesexposed to high temperature and radiation. Ultrasonic wavesare generated using an Nd:Yag laser beam, which generatesa localized thermal expansion on the pipe surface, and thecorresponding ultrasonic responses are measured using a laserinterferometer, which measures the out-of-plane velocity ofthe surface. The laser beams from the sources are guidedand transmitted through optic fibers to target excitation andsensing points on the pipe surface. The measurement systemis especially designed to operate under radioactive and high-temperature environments inside NPPs.
This paper is organized as follows. First, the authorsdescribe the optical fiber-guided laser ultrasonic systemdesigned for the generation and measurement of ultrasonicwaves. Application tests under high temperature and radiationenvironment are then followed to check the durability of thesystem under the operational condition of NPPs. Finally, thispaper concludes with a summary and future work.
2. Optical fiber-guided laser ultrasonic generationand measurement system
Figure 1 shows a schematic diagram of the proposedoptical fiber-guided laser ultrasonic system. For the ultrasonicgeneration on a pipe surface, a pulse laser is connected toone end of an optical fiber, transmitted through the opticalfiber with little power loss, and emitted at the other end of thefiber. This end of the optical fiber is fixed to a pipe surfaceusing a focusing module and stainless steel strip. The focusingmodule is used to focus a laser beam for the proper ultrasonicgeneration and a stainless steel strip is used to firmly fixthe focusing module to the pipe surface. The correspondingultrasonic response is measured at another point using a fiberinterferometer connected to a separate optical fiber. Here, oneend of the fiber is attached to the pipe surface similar to theone used for ultrasonic excitation.
Figure 2 shows the detailed configuration of the ultrasonicgeneration segment composed of an optical fiber, laser source,connectors, microlens arrays and convex lens. When a high-power laser beam is exposed to a solid surface, ultrasonicwaves are generated by thermoelastic expansion of the exposedsurface (Scruby and Drain 1990). Here, the power level,laser pulse duration and laser beam size need to be carefullycontrolled because the laser power density above a certainthreshold will cause ablation on the surface. Ablation generallyoccurs in a metallic structure with power density above107 W cm−2, but this value varies depending on actual materialproperties, surface condition and especially absorption ratio(Pierce et al 1998).
In this study, a 532 nm Nd:YAG pulse laser (Quantel,Brilliant Ultra) is used. The maximum output energy per unitpulse, the pulse duration and the repetition rate of the laserbeam are 30 mJ, 8 ns and 20 Hz, respectively. The powerdensity incident on the pipe was set to be 0.75 MW cm−2 at theemitting end of the fiber to avoid ablation. A metal-tubed stepindex multimode fiber with 1000 μm core diameter (Thorlabs,FT1000EMT) is used to carry the laser beam from the lasersource to the target point on the pipe surface. This specifictype of fiber is selected because it can operate under hightemperature and radioactive environment inside NPPs.
Note that the high-power laser beam needs to be properlyfocused on the multimode fiber to minimize power loss andfiber damage. In this regard, the profile of the laser beamintensity across the fiber cross section is changed from aGaussian distribution to a flat top distribution using twomicrolens arrays (Edmund, P64-478). Then, the laser beamis focused using an N-BK7 plano-convex lens (Thorlabs,
2
Meas. Sci. Technol. 24 (2013) 085003 J Yang et al
Figure 2. Configuration of the optical fiber-guided laser ultrasonic generation segment (units: mm).
Figure 3. Configuration of the optical fiber-guided laser ultrasonic measurement segment (units: mm).
LA1805-YAG) with a focal length of 50 mm before enteringthe SMA connector. Finally, the laser beam from the otherSMA travels through a metal tube and creates a laser spot of7–8 mm diameter on the pipe surface. A stainless steel strip isused to fix the focusing module to the specimen.
Figure 3 shows the detailed configuration of the opticalfiber-guided laser ultrasonic measurement segment. A fiber-coupled laser interferometer (Polytec, OFV-551) is used tomeasure the out-of-plane velocity generated by ultrasonicwaves propagating on the pipe surface. The working principleof ultrasonic measurement using the laser interferometeris based on the Doppler effect (Kundu 2004). The laserinterferometer launches a continuous laser beam to a targetobject and then receives a reflected beam from the objectsurface. Here, the optical phase difference between the incidentand reflected laser beams becomes proportional to the out-of-plane velocity of the target surface. Therefore, by tracking this
phase shift, ultrasonic wave velocities on the pipe surface canbe measured. In this study, a He–Ne continuous laser sourcewith 633 nm wavelength, 1 mW maximum power and 16 μmbeam size is used for laser interferometry.
As for the optical fiber in the measurement segment,a single-mode fiber is used instead of the multimode fiberused in the excitation segment. The power required for themeasurement laser beam is less than for the pulsed excitationlaser beam and the single-mode fiber also has narrowermodal dispersion characteristics. Furthermore, a polarization-maintaining fiber is used to maintain the polarization of themeasurement laser beam and reduce the polarization-inducedpower loss that governs the signal-to-noise ratio of responsesignals (Castellini et al 2006).
The sensor head of the laser interferometer is insertedinto a focusing module to enhance focusing and protect thesensor head. A fused-silica focusing lens with a focal length
3
Meas. Sci. Technol. 24 (2013) 085003 J Yang et al
Figure 4. Setup for high-temperature experiments.
of 15 mm is installed close to the target surface. Since the targetpipe surface is exposed to a high-temperature environment, aheat absorption filter using KG1 glass is attached close tothe sensor head to protect the sensor head from infrared heatradiation. The fixture strip and metal tube identical to the onesin the wave generation segment are used to fix the fiber to thepipe.
Since the source of the ultrasonic wave in this study is thepulse-shaped laser beam, the generated ultrasonic waves have abroadband frequency spectrum. For the further signal analysis,however, this study observes only the frequency componentsup to 300 kHz, considering that the frequency range upto hundreds of kilohertz is the main interest of ultrasonicinspection. The signals at the higher frequency range have alow amplitude, low signal-to-noise ratio and multimode nature,which limit further applications of ultrasonic inspection. Thesignals at a much lower frequency level, on the other hand,are highly influenced by low-frequency structural vibrationsregardless of damage-sensitive ultrasonic signal changes(Giurgiutiu 2008). One more practical reason for the band-pass filtering is the fact that the velocity decoder of the fiberinterferometer (Polytec, VD-07) used in this study measuresthe frequency component of the response signal up to 350 kHzby the device specification.
3. Effects of high temperatures on the ultrasonicwave measurement
First, the effect of temperature on the proposed laserultrasonic system is investigated. For typical NPPs, thenominal operational temperature of the primary coolantsystem can increase up to 350 ◦C near the reactor and theoperating temperature of the secondary piping system is
maintained around 100–200 ◦C. Figure 4 shows the overallconfiguration of the high-temperature experiment conductedin this study. The temperature experiment is conducted usingan especially designed temperature controller (SKI Ltd) usedby a manufacturer of insulation materials for NPP pipes. Thecontroller has a built-in pipe and a magnetic heater inside thepipe to heat up the pipe. The axial length, outer diameter andthickness of the pipe are 600, 114.3 and 3 mm, respectively.
The peak power of the Nd:YAG laser was set to0.75 MW cm−2 at the emitting end of fiber and thecorresponding responses were measured by the fiberinterferometer at a sampling rate of 5.12 MHz. The responsesignals were measured 512 times under identical conditions,averaged in the time domain to improve the signal-to-noiseratio. The distance between the ultrasonic generation andsensing points is set to be 300 mm.
Figure 5 shows the ultrasonic response signals obtainedunder varying temperature conditions. The temperature isincreased from room temperature to 340 ◦C and the responsesignals are measured from 200 to 340 ◦C with a 10 ◦Cincrement. Each signal is band-pass-filtered with low and highcutoff frequencies of 200 and 300 kHz. As the temperatureincreases, the first arrival peak of the signal is decreased anddelayed in the time domain. The amplitude and arrival time ofthe response signal are 0.6948 mV and 0.1352 ms at 260 ◦C andchange to 0.6432 mV and 0.136 ms at 300 ◦C, and 0.5755 mVand 0.137 ms at 340 ◦C, respectively. It is speculated that theattenuation and time delay are mainly caused by the reductionof Young’s modulus due to temperature increase (Raghavanand Cesnik 2008).
The experimental results presented here highlight theimportance of temperature compensation at the actual damagedetection step. Several techniques such as optimal baseline
4
Meas. Sci. Technol. 24 (2013) 085003 J Yang et al
Figure 5. Comparison of the ultrasonic response signals obtained from varying pipe temperatures.
subtraction (Croxford et al 2007), baseline stretching (Liuand Michaels 2005) and data normalization (Oh and Sohn2009, Lim et al 2011) have been proposed to minimize falsealarms due to temperature changes. A combined optimalbaseline subtraction and stretch method (Clarke et al 2009)compensates the effect of temperature on an initial baselinesignal and the baseline signal is subtracted from a test signalto isolate signal changes only caused by damage. Similarly,the damage detection using data normalization is conductedby identifying a new dataset that significantly deviates froma pool of multiple baseline datasets. On the other hand, abaseline-free technique, which identifies damage without anybaseline data obtained from the pristine condition of the teststructure, extracts a damage feature based on the premise thatthe axisymmetric nature of ultrasonic wave signals obtainedfrom a pipe breaks down only when structural damage existsand the structural axisymmetry of the pipe is independent ofpipe temperature (Lee et al 2012).
4. Effects of gamma radiation on the ultrasonic wavemeasurement
Next, the effect of gamma radiation on the proposed laserultrasonic system is investigated. For typical NPP containmentbuildings, the normal gamma radiation accumulated doselevel is about 500 kGy for 40 years and the dose ratethat is emitted from the radiation source is approximately103–107 Gy h−1 (Francis et al 1998). But only some of thisactually reaches a target structure. Optical fibers exposed togamma radiation exhibit effects such as radiation-inducedabsorption (RIA), radiation-induced luminescence (RIL),change of the waveguide refractive index, etc (Friebele 1979,Friebele et al 1984). Among them, RIA and RIL effectscontribute to the degradation of the signal-to-noise ratio of thesignals transmitted over the optical fiber guide (Gill et al 1997,Regnier et al 2007). Therefore, the current study investigatesthe effect of radiation on optical fibers by examining the
change of ultrasonic time signals after exposing optical fibersto gamma radiation.
Figure 6 shows the gamma radiation facility at KoreaAtomic Energy Research Institute (KAERI). Radiationexposure is performed using a Co-60 (364 000 Ci) source.Principal gamma radiation energy from the radiation sourcevaries from 1.17 to 1.33 MeV (1.25 MeV on average). Theradiation dose rate of this facility is in the range of 0.4–20 kGy h−1.
Radiation experiments are performed using six fibers(Thorlabs, FT1000EMT). The target dose rate is set to 5, 10and 20 kGy h−1, and the total dose to 62.5 kGy (typical five-year dose) and 125 kGy (typical ten-year dose), respectively.The radiation dose is gauged by using dosimeters placed nearthe specimen inside the radiation facility. Note that only opticalfibers are exposed to gamma radiation in this experimentsince the other hardware components are placed outside theradiation containment. A stainless steel pipe with an axiallength of 1000 mm, outer diameter of 165 mm and thicknessof 3 mm is used for this study. The distance between theultrasonic generation point and the measurement point is setto be 400 mm. The rest of the experimental setup is identicalto the previous experiment and the experiment is conducted atroom temperature.
Figure 7 shows representative ultrasonic response signalswhen an optical fiber is irradiated for a total of 125 kGy gammaradiation for 6 h 15 min at a 20 kGy h−1 dose rate. As the opticalfiber is exposed to more radiation, the amplitude of the firstarrival peak is reduced as a result of optical fiber hardening(Heschel et al 2002). A quantitative summary of the peakamplitude reduction is provided in table 1 as a function of theradiation dose rate and radiation time. Note that the waveformof the first arrival peak is reasonably well preserved after theband-pass filtering throughout the experiments.
Figure 8 shows ultrasonic responses after repeatedradiation tests. Each radiation test lasts for 12 h 30 minwith a dose rate of 10 kGy h−1. As the radiation cycle
5
Meas. Sci. Technol. 24 (2013) 085003 J Yang et al
(a)
(b)
(c)
(d)
(e)
Figure 6. Gamma radiation facility at KAERI. (a) The source hoist cable places the gamma ray source in and out of the pool. (b) The sourceleg guide cables stabilize the gamma ray source pathway. (c) The source shroud protects the gamma ray source. (d) Test samples are placedon the table/testing stand. (e) The source pool is used for the water storage of the gamma ray source.
Figure 7. Comparison of ultrasonic responses obtained before and after gamma radiation (20 kGy h−1, 125 kGy).
Table 1. Amplitude attenuation of the first arrival peak.
Attenuation of the first arrival peakTest radiation time amplitude w.r.t. the initial peak value
Dose rate (equivalent normal operation time) before radiation
5 kGy h−1 12.5 h (five years) 0.4716 dB25 h (ten years) 0.8621 dB
10 kGy h−1 6.25 h (five years) 1.1981 dB12.5 h (ten years) 0.7756 dB
20 kGy h−1 3.125 h (five years) 1.5109 dB6.25 h (ten years) 1.2840 dB
6
Meas. Sci. Technol. 24 (2013) 085003 J Yang et al
Figure 8. The effect of repeated gamma radiation exposures on ultrasonic responses (10 kGy h−1).
increases, the amplitude of the first arrival peak decreases withlittle waveform change. The peak amplitude is reduced from0.84 mV to 0.70, 0.63 and 0.54 mV after three consecutiveradiation cycles, respectively.
5. Conclusions
This study develops an optical fiber-guided laser ultrasonicsystem that can be applied for real environmental conditionsof nuclear power plants (NPPs). An optical fiber which isespecially designed with a focusing lens and a fixture is usedto deliver a pulse laser to an excitation point for ultrasonicwave generation. Another optical fiber is connected between afiber vibrometer and a measurement point for ultrasonic wavesensing with a similar focusing lens and a fixture. To verifythe validity of the proposed system, this study has performedseveral experimental tests under high-temperature and high-radiation environments. The experimental results confirm thatthe proposed system can operate under (1) temperature rangesup to 340 ◦C, which is the highest limit in the primary systemof NPPs, and (2) radiation ranges up to 20 kGy, which isequivalent to the radiation dose for the operation period often years. Follow-up studies are underway to evaluate theperformance of the proposed system. First, a robust damagedetection algorithm is being built considering signal changesunder varying temperature and radiation conditions in orderto minimize false alarms due to these changes. Also, anoptical switching device is being developed so that selectiveultrasonic generation and measurement can be achieved atmultipoints.
Acknowledgments
This work was supported by the Radiation TechnologyProgram (2010-0020010) of National Research Foundation(NRF) of Korea funded by Ministry of Education, Scienceand Technology (MEST) and by the Innovations in Nuclear
Power Technology of the Korea Institute of EnergyTechnology Evaluation and Planning (KETEP) grant fundedby the Korea Government Ministry of Knowledge Economy(2010T100101057) in Korea.
References
Adams D E 2007 Health Monitoring of Structural Materials andComponents (Chichester: Wiley)
American Nuclear Society 2011 Fukushima Daiichi: ANSCommittee Report (LaGrange Park, IL: American NuclearSociety)
Blitz J and Simpson G 1996 Ultrasonic Methods of Non-DestructiveTesting (Cambridge: Chapman and Hall)
Carts L 1995 Nondestructive Testing: Radiography, Ultrasonics,Liquid Penetrant, Magnetic Particle, Eddy Current (MaterialsPark, OH: ASM International)
Castellini P, Martarelli M and Toamasini E P 2006 Laser Dopplervibrometry: development of advanced solutions answering totechnology’s needs Mech. Syst. Signal Process. 20 1265–85
Clarke T, Cawley P, Wilcox P D and Croxford A J 2009 Evaluationof the damage detection capacity of a sparse-array guided-waveSHM system applied to a complex structure under varyingthermal conditions IEEE Trans. Ultrason. Ferroelectr. Freq.Control 56 2666–78
Croxford A J, Wilcox P D, Drinkwater B W and Konstantinidis G2007 Strategies for guided wave structural health monitoringProc. R. Soc. Lond. A 463 2961–81
Francis B et al 1998 Radiation-hardness of fiber-optic sensors formonitoring and remote handling applications in nuclearenvironments Proc. SPIE 3538 28–39
Friebele E J 1979 Optical fiber waveguides in radiationenvironments Opt. Eng. 18 552–61
Friebele E J, Askins C G, Gingerich M E and Long K J 1984Optical fiber waveguides in radiation environments: II Nucl.Instrum. Methods Phys. Res. B 1 355–69
Gill K, Grabit R, Persello M, Stefanini G and Vasey F 1997 Gammaand neutron radiation damage studies of optical fibresJ. Non-Cryst. Solids 216 129–34
Giurgiutiu V 2008 Structural Health Monitoring with PiezoelectricWafer Active Sensors (Boston, MA: Elsevier)
Hashemian H M 2010 Aging management of instrumentation &control sensors in nuclear power plants Nucl. Eng. Des.240 3781–90
Heschel H, Kohn O and Weinand U 2002 A new radiation hardoptical fiber for high-dose values IEEE Trans. Nucl. Sci.49 1432–8
Inman D J 2005 Damage Prognosis for Aerospace, Civil andMechanical Systems (Chichester: Wiley)
Kundu T 2004 Ultrasonic Nondestructive Evaluation: Engineeringand Biological Material Characterization (London: CRCPress) pp 121–38
Lee H, Yang J and Sohn H 2012 Baseline-free pipeline monitoringusing optical fiber-guided laser ultrasonics Struct. HealthMonit. 11 684–95
Lim H J, Kim M K, Sohn H and Park C Y 2011 Impedance baseddamage detection under varying temperature and loadingconditions NDT & E Int. 44 740–50
Liu Y and Michaels J E 2005 A methodology for structural healthmonitoring with diffuse ultrasonic waves in the presence oftemperature variations Ultrasonics 43 717–31
Oh C K and Sohn H 2009 Damage diagnosis under environmentaland operational variations using unsupervised support vectormachine J. Sound Vib. 325 224–39
Pierce S G, Culshaw B and Shan Q 1998 Laser generation ofultrasound using a modulated continuous wave laser diodeAppl. Phys. Lett. 72 1030–2
Raghavan A and Cesnik C E S 2008 Effects of elevated temperatureon guided-wave structural health monitoring J. Intell. Mater.Syst. Struct. 19 1383–98
Regnier E, Flammer I, Girard S, Gooijer F, Achten F and Kuyt G2007 Low-dose radiation-induced attenuation at InfraRedwavelengths for P-doped, Ge-doped and pure silica-coreoptical fibres IEEE Trans. Nucl. Sci. 54 1115–9
Scruby C B and Drain L E 1990 Laser Ultrasonics: Techniquesand Applications (New York: Taylor and Francis)pp 76–85
Sohn H, Farrar C R, Hemez F M, Czarnecki J J, Shunk D D,Stinemates D W and Nadler B R 2003 A review of structuralhealth monitoring literature: 1996–2001 Los Alamos NationalLaboratory Report LA-13976-MS (Los Alamos: Los AlamosNational Laboratory)
US Energy Information Administration 2008 International EnergyOutlook (Washington DC: Department of Energy)
